An SOS-regulated operon involved in damage-inducible mutagenesis in Ca
http://www.100md.com
《核酸研究医学期刊》
Department of Microbiology, Institute of Biomedical Sciences, University of S?o Paulo S?o Paulo, SP 05508-900, Brazil
*To whom correspondence should be addressed at Department of Microbiology, Institute of Biomedical Sciences, University of S?o Paulo, Avenue Professor Lineu Prestes, 1374, S?o Paulo, SP 05508-900, Brazil. Tel: +55 11 3091 7499; Fax: +55 11 3091 7354; Email: cfmmenck@usp.br
ABSTRACT
DNA polymerases of the Y-family, such as Escherichia coli UmuC and DinB, are specialized enzymes induced by the SOS response, which bypass lesions allowing the continuation of DNA replication. umuDC orthologs are absent in Caulobacter crescentus and other bacteria, raising the question about the existence of SOS mutagenesis in these organisms. Here, we report that the C.crescentus dinB ortholog is not involved in damage-induced mutagenesis. However, an operon composed of two hypothetical genes and dnaE2, encoding a second copy of the catalytic subunit of Pol III, is damage inducible in a recA-dependent manner, and is responsible for most ultraviolet (UV) and mitomycin C-induced mutations in C.crescentus. The results demonstrate that the three genes are required for the error-prone processing of DNA lesions. The two hypothetical genes were named imuA and imuB, after inducible mutagenesis. ImuB is similar to proteins of the Y-family of polymerases, and possibly cooperates with DnaE2 in lesion bypass. The mutations arising as a consequence of the activity of the imuAB dnaE2 operon are rather unusual for UV irradiation, including G:C to C:G transversions.
INTRODUCTION
DNA damage-induced mutagenesis is, in a major extent, an active process that requires specialized DNA polymerases able to perform translesion synthesis (TLS). In Escherichia coli, three SOS-regulated DNA polymerases are involved in damage tolerance: Pol II, encoded by the polB gene, Pol IV and Pol V. The major role of Pol II is the rescue of stalled replication forks (1), although it can bypass some types of DNA damage (2). dinB (also called dinP) encodes a member of the Y-family of DNA polymerases, DNA Pol IV (3,4). This enzyme is responsible for untargeted SOS mutagenesis of phage , and induces frameshift mutations when overexpressed (5). More recently, dinB has been associated with the phenomenon of mutation under stress conditions, or adaptive mutagenesis .
In E.coli, ultraviolet (UV) and chemical mutagenesis is heavily dependent on the SOS-regulated umuDC operon (7,8). The UmuC protein has been identified as a DNA polymerase (Pol V), a distributive enzyme able to perform TLS and highly inaccurate even when using undamaged templates (9,10), which interacts with the processed form of the co-expressed protein UmuD. Both in vivo and in vitro experiments have shown that efficient TLS requires the complex, RecA, Ssb, the catalytic subunit of DNA Pol III encoded by the dnaE gene, and the ? clamp .
DNA polymerase III is a multiprotein complex responsible for the replication of the bacterial chromosome. Although the Y-family DNA polymerases, specialized in TLS, seem to be ubiquitous throughout nature, it has been shown that the subunit of Pol III, which bears the catalytic polymerase activity, is also able to perform TLS. In Bacillus subtilis, the dnaE gene is SOS regulated and the purified enzyme can perform TLS in vitro, an activity also shown for the Streptococcus pyogenes protein (12,13). Depletion of B.subtilis DnaE inhibits UV-induced mutagenesis, although it is difficult to ascertain whether this requirement reflects the need for this enzyme to extend the bypass reaction performed by Y-family DNA polymerases, or results from actual lack of lesion bypass (13).
It has recently been shown that a second copy of the dnaE gene mediates SOS mutagenesis in Mycobacterium tuberculosis (14). The expression of this gene is regulated by the SOS response in this organism, being induced by DNA damaging agents in a recA-dependent manner (15,16). The dnaE gene is presented in duplicate in several bacterial genomes, most frequently in a potential operon with two hypothetical genes, one of which is a dinB-related one. This widespread operon has been shown to be regulated by LexA in Pseudomonas putida, indicating that it is involved in DNA damage response (17). This operon is present in several -proteobacteria genomes completely sequenced to date, including C.crescentus (18).
In C.crescentus, a recA-dependent inducible repair system was demonstrated through the existence of Weigle reactivation of irradiated phage (19). This organism is also mutable by UV radiation, suggesting the existence of error-prone pathways, although umuDC orthologs are absent from its genome (18,19). In this work, we investigated the involvement of the dnaE2-containing gene cassette and the dinB ortholog in damage-induced mutagenesis, detected in rifampicin resistance assays. The results show that the dnaE2 operon is upregulated in response to DNA damage in a recA-dependent manner, and demonstrate that all three genes that compose this expression unit are involved in UV and mitomycin C-induced mutagenesis. On the other hand, dinB is not necessary for mutagenic repair. We propose to name the genes co-expressed with dnaE2, after inducible mutagenesis, as imuA and imuB, and show that the imuAB/dnaE2-dependent mutational spectra of UV-induced lesions is different from the prototypical E.coli one, including a high proportion of G:C C:G transversions.
MATERIALS AND METHODS
Bacterial strains and plasmids
Bacterial strains and plasmids used in this study are shown in Table 1. C.crescentus strains were grown in PYE medium (20) at 30°C with constant shaking. Plasmids were introduced into C.crescentus by conjugation with E.coli S17-1 strain. When appropriated, the culture medium was supplemented with kanamycin (50 μg/ml), nalidixic acid (25 μg/ml), spectinomycin (50 μg/ml) or tetracycline (1 μg/ml). E.coli strain DH10B (Invitrogen, CA) was used for cloning purposes. E.coli was grown at 37°C in Luria–Bertani medium supplemented with ampicillin (100 μg/ml), kanamycin (50 μg/ml) or tetracycline (15 μg/ml) when necessary.
Table 1 Bacterial strains and plasmids used in this study
Genetic disruptions were achieved using the pNPTS138 vector, and selecting for two consecutive recombination events. First, Kanr conjugants of C.crescentus were selected in the screening for the first recombination event. Loss of the plasmid after the second recombination was selected in PYE media containing 3% sucrose. When appropriated, a concomitant Spcr selection was also applied. Strains generated in this way were analyzed by Southern blot to confirm gene disruptions. The construction of gene targeting plasmids was performed with PCR products amplified with the primers shown in Supplementary Table 1S. These products were cloned in the pGEM-T easy vector and sequenced to assure sequence integrity. For disruption of recA, fragments amplified with primers recaA x recaB, and recaC x recaD were fused in the pNPTS138 vector using the restriction sites introduced, generating a 1.8 kb fragment containing the first 81 bp fused to the last 75 bp of recA plus flanking regions, leading to an in-frame deletion of recA. In-frame deletions were constructed in the same way for imuA and imuB using, respectively, primers CC3213A to D and CC3212A to D. imuA disruption resulted in a deletion containing only the first 36 and the last 54 bp of the gene, and imuB disruption resulted in a deletion containing only the first 63 and the last 81 bp of the gene. dnaE2 disruption was performed by the insertion of the specr cassette (21) in the natural EcoRI site of a 1629 bp fragment of this gene, amplified with the dnaE2re and dnaE2fo primers. The insertion in the EcoRI site resulted in a predicted truncation at position 598 of the protein, leading to loss of the last 485 amino acids. dinB knock-out was constructed by insertion of the specr cassette in the EcoRV site of a 1.2 kb fragment, amplified with primers dinbfo and dinbre. This insertion resulted in a predicted truncation at position 167 of the 419 amino acid DinB protein.
PCRs were also carried out to amplify promoter regions of uvrA and imuA genes. The primers puvrafo and puvrare were used to obtain a fragment ranging from –492 to +54 of the start codon of uvrA . Primers p3213fo and p3213re amplified a fragment ranging from –513 to +74 of the imuA start codon. Both fragments were cloned in the pRKlacZ290 vector using the restriction sites introduced in the primers.
In order to promote complementation of the phenotypes of the imuA strain, a full-length gene including the promoter region was amplified with the 3213A and 3213D oligonucleotides, and cloned in the low copy number pMR20 vector, using the restriction sites introduced in the primers.
Survival and mutagenesis experiments
Caulobacter crescentus strains were grown until mid-log phase (5 x 108 cells/ml), and then submitted to the different treatments. A 5 ml batch of culture was UV-irradiated with a Philips TUV 15W/G15T8 germicidal lamp (mainly 254 nm) in Petri dishes. Aliquots were removed before and after irradiation for serial dilutions and plating in PYE medium, for the determination of viable cell counts after 48 h incubation at 30°C. For mutagenesis experiments, a 200 μl aliquot of irradiated cells was inoculated in 1 ml of PYE liquid medium and cultivated overnight, to allow mutation fixation. Then, the cultures were submitted to serial dilutions and plating on PYE medium for viable cell counts. The cultures were also plated in PYE medium containing 100 μg/ml rifampicin to score Rifr cells. Mutation frequencies were calculated by dividing the numbers of mutants by the estimated total number of cells analyzed in the experiments. Mitomycin C (mit C) survival and mutagenesis experiments were conducted in a similar manner, with drug treatments performed in 2 ml of culture for 1 h at 30°C. For mit C mutagenesis, cells were centrifuged and resuspended in PYE before inoculating for overnight incubation, to remove the drug from the medium.
Determination of the mutation spectra in the rpoB gene
Rifampicin-resistant clones, obtained after UV irradiation in the experiments described above, were cultivated overnight for the extraction of chromosomal DNA. The primers rpoB2fo and rpoB2re (Supplementary Table 1S) were used in PCRs designed for the amplification of the region II of rifampicin resistance. PCR products were purified and sequenced in both strands using rpoB2fo and rpoB2re primers with the Big Dye Terminator Cycle Sequencing kit and analyzed in the ABI Prism 3100 Genetic Analyzer (Applied Biosystems, CA). Only independent mutations from each experiment were scored.
RT–PCR experiments and ?-galactosidase assays
The operon organization of genes CC3211, CC3212 and CC3213 was assayed through RT–PCR experiments. The primers 1112A and 1112B were used to check imuB/dnaE2 co-expression, and primers 1213A and 1213B were used to check imuA/imuB co-expression. RNA from exponentially growing cells was extracted with Trizol? reagent (Invitrogen), and treated with DNAseI to eliminate contaminant DNA. RT–PCR experiments were carried out with 300 ng of RNA using the Superscript One Step kit (Invitrogen), in the following cycling conditions: 50°C for 20 min; 94°C for 2 min; 40 cycles of 95°C for 1 min; 61°C for 30 s; 72°C for 50 s; followed by incubation at 72°C for 7 min and cooling to 4°C. A control for DNA contamination of samples was performed with PCR lacking reverse transcriptase.
Measurements of promoter activity with lacZ transcriptional fusions were performed after UV irradiation, in the same conditions used for survival experiments (see above), allowing cells to recover in PYE medium for 90 min after treatment. After recovery, cells were assayed for ?-galactosidase activity as described previously (22).
Real-time analysis of gene expression
Relative expression of imuA, imuB, dnaE2 and dinB after UV irradiation was determined with quantitative PCR experiments. An aliquot of 2 μg of RNA pre-treated with DNAse I were used as template for total cDNA synthesis in 20 μl reactions with random hexamers using the Superscript First-Strand Synthesis System for RT–PCR (Invitrogen). For real-time PCR, an amount of cDNA corresponding to 25 ng of input RNA was used in each reaction. Reactions were performed with the SYBR Green PCR Master Mix (Applied Biosystems), and analyzed in the ABI 7500 Real-Time System. Relative expression levels were calculated as described previously (23), using the rpoD gene as the endogenous control. Primers used in this analysis are listed in Supplementary Table 1S.
Phylogenetic analyses
Protein sequences of genes CC3211, CC3212 and CC3213 were obtained from the National Center of Biotechnology Information GenBank database (www.ncbi.nlm.nih.gov/Entrez/). The accession numbers are respectively NP_422005 , NP_422006 and NP_422007 . Orthologous protein sequences from complete genomes were obtained by BLAST analyses (24). The lack of orthologs in several genomes was ascertained by exhaustive BLAST analysis.
For phylogenetic analysis, the protein sequences were aligned using the ClustalX multiple sequence alignment program (25) with manual adjustment using Genedoc v2.6.001 (26) (www.psc.edu/biomed/genedoc). Only unambiguously aligned positions (excluding poorly conserved and gap regions) were used. Phylogenetic trees were generated for each group of protein homologs from sequence alignments using the PHYLIP program version 3.5 (27). Distance analyses were performed using the neighbor-joining method in PHYLIP, with the distance PAM matrix model (28). Bootstrap support (resampled 1000 times) was calculated, and strict consensus trees constructed. Only bootstrap values >50% are shown. The consensus trees obtained were viewed through TreeView software (29). The same set of prokaryotic species was used in all analyses, and includes CC3211 (dnaE2) and CC3212 (imuB) orthologs from fully sequenced genomes and E.coli proteins.
RESULTS
Genomic context and phylogenetic analyses
The CC3211 gene of C.crescentus encodes a second copy of the dnaE gene, hereafter referred to as dnaE2. dnaE2 is in a conserved organization with two genes encoding hypothetical proteins, CC3212 (imuB) and CC3213 (imuA), similar to other proteobacteria (Figure 1). This operon organization is eventually associated with a duplication of lexA, such as in P.putida. However, this is not the case for C.crescentus and related bacteria, which have only one copy the lexA gene (30). It was previously shown that the operon organization of imuAB and dnaE2 is conserved in all genomes of proteobacteria containing these genes (17). However, other bacterial species, such as several Actinomycetales and the planctomycete Rhodopirellula baltica, also possess a dnaE2 gene, although in different genomic contexts (Figure 1). A total of six different gene arrangements were found in bacterial genomes containing dnaE2. Remarkably, the large majority of organisms presenting dnaE2 also possess an imuB ortholog, although not necessarily in an operon organization. In R.baltica, imuA and imuB form a potential operon, which does not include dnaE2, located apart in the genome. The M.tuberculosis imuB gene, encoded by the Rv3394c ORF, is in an operon with Rv3395c, a putative RecA-like protein. Most interestingly, both Rv3395c and Rv3394c are damage inducible in M.tuberculosis, and the promoter of Rv3395c has been shown to contain a Mycobacterial SOS box (14,16). A similar picture can be observed in the genomes of other Actinomycetales, except Propionibacterium acnes, where the Rv3395c ortholog forms a putative operon with both imuB and dnaE2. Although proteins with significant similarity to ImuA cannot be found in Actinomycetales, the genomic context analysis suggests that the RecA-like protein encoded by M.tuberculosis Rv3395 may represent a functional counterpart of ImuA in these organisms. It is evident that several rearrangements of the genomic organization of dnaE2 and imuB have taken place during evolution, but notably these genes always co-exist in bacterial genomes. These data indicate that the imuB and dnaE2 genes compose a highly conserved co-evolving genetic system, which is likely to be involved in DNA damage response.
Figure 1 Prototypical genomic organizations of dnaE2 and accompanying imuA and imuB genes. The same organization found in C.crescentus is also present in -proteobacterial genomes, Pseudomonas aeruginosa, Vibrio vulnificus, Vibrio parahaemolyticus, Ralstonia solanacearum, Bordetella bronchiseptica and Bordetella parapertussis. A duplication of lexA preceding the operon, as found in P.putida, is also present in Methylococcus capsulatus, Pseudomonas syringae, Xanthomonas campestris and Xanthomonas axonopodis. The M.tuberculosis gene arrangement is conserved in all species of the genera Mycobacterium and Corynebacterium, and in Nocardia farcinica. Small boxes represent known SOS operators, and interrupted lines indicate that the genes are not contiguous in the genome. Loci numbers as provided by the genome annotation are indicated above, the arrows representing genes, and our annotation is given inside the arrows. Different original gene annotations are indicated in parentheses.
Phylogenetic analyses of the DnaE proteins of the microorganisms presenting such duplication indicate that this is most probably a very ancestral duplication event, present in the vast majority of fully sequenced genomes from -proteobacteria, but also found in other proteobacteria and in Actinomycetales (Figure 2). The branch of DnaE includes the E.coli protein and the product of the CC1926 gene, which seems to be involved in DNA replication in C.crescentus (31). DnaE2 orthologs clearly form a distinct branch, including the DnaE2 from C.crescentus. The prototypical member of this branch is the M.tuberculosis protein Rv3370c. This protein has been found to be damage inducible, and implicated in cell survival and mutagenesis after UV treatment (14).
Figure 2 Phylogenetic relationships between DnaE and DnaE2. Phylogenetic analyses included DnaE and DnaE2 proteins of all fully sequenced genomes where such duplication could be found. Only M.tuberculosis and Mycobacterium bovis were included for simplification, although the duplication can be found in several other Mycobacterial genomes. The names of the ORFs and species name abbreviations shown are the same adopted in the genome annotation of each organism. Abbreviations of organism names are as follows: AGR, A.tumefaciens str. C58; b, E.coli K12; BB, B.bronchiseptica RB50; bll, Bradyrhizobium japonicum USDA 110; BMA, Burkholderia mallei ATCC 23344; BMEI, Brucella melitensis 16M; BPP, B.parapertussis 12822; BPSL, Burkholderia pseudomallei K96243; BR, Brucella suis 1330; CC, C.crescentus CB15; CE, Corynebacterium efficiens YS-314; DIP, Corynebacterium diphtheriae NCTC 13129; Mb, M.bovis AF2122/97; MCA, M.capsulatus str. Bath; mll and mlr, Mesorhizobium loti MAFF303099; NCgl, Corynebacterium glutamicum ATCC 13032; Nfa, N.farcinica IFM 10152; PA, Pseudomonas aeruginosa PAO1; PP, P.putida KT2440; PPA, P.acnes KPA171202; PSPTO, P.syringae pv. tomato str. DC3000; RB, R.baltica SH 1; RPA, Rhodopseudomonas palustris CGA009; Rsc and Rsp, R.solanacearum GMI1000; Rv, M.tuberculosis H37Rv; SAV, S.avermitilis MA-4680; Smc and Sma, S.meliloti 1021; VP, V.parahaemolyticus RIMD 2210633; VV, V.vulnificus YJ016; XAC, X.axonopodis pv. citri str. 306; and XCC, Xanthomonas campestris pv. campestris str. ATCC 33913. The numbers indicate the bootstrap values >50%.
The CC3213 ortholog from P.putida, PP3117, was annotated as sulA2 (32). However, PSI-BLAST analyses of the protein encoded by CC3213 shows only very low levels of similarity to SulA (24), and also some similarity to RecA/RadA recombinases (data not shown). A phylogenetic tree of this protein and its orthologs has been presented previously (17), but the levels of protein similarities between CC3213 orthologs and known proteins, such as SulA and RecA, are too low to precisely ascribe any protein function. We tested the filamentaion in both the wild type and the derivative CC3213 deletion strain in several conditions by microscopic examination. C.crescentus shows a filamentous morphology after prolonged exposure to mit C, and the deletion of CC3213 did not affect this cellular response, indicating that this gene is not involved in cell division suppression (Supplementary Figure 1S). Therefore, we propose to name the CC3213 gene imuA, after inducible mutagenesis, given the functional characterization conducted (see below).
Similarly, CC3212 orthologs identified by BLAST analysis present diverse annotations: the P.putida gene PP3118 was annotated as dinP (32), and the PPA1651 gene of P.acnes was annotated as umuC (33), while most other orthologs (including CC3212 itself) were annotated as genes with unknown function. In this case, phylogenetic analysis was conducted including all fully sequenced genomes where CC3212 orthologs were found, for comparison between these proteins and DinB, the closest related protein of known function. All organisms with a CC3212 ortholog also have one or more typical dinB genes. The product of CC3212 and its orthologs form a branch that is distinct from the DinB proteins of the same organisms, indicating that this is a divergent DinB/UmuC homolog (Figure 3). The only possible exception is the SAV6556 protein of Streptomyces avermitilis, which cannot be surely assigned as ImuB through phylogenetic analyses, although it is present in a putative operon with dnaE2 (Figure 1). Protein similarity comparison among these proteins reinforces the difference between ImuB and DinB, e.g. E.coli DinB and UmuC proteins share 45% identity, while C.crescentus DinB and the ImuB proteins share only 35% identity. Therefore, we concluded that the CC3212 gene product is a member of a distinct branch in the UmuC superfamily of proteins, and not a DinB ortholog. We named this gene imuB, in accordance with the nomenclature adopted for CC3213.
Figure 3 Phylogenetic relationships between ImuB and DinB. Phylogenetic analyses included ImuB and DinB proteins of all fully sequenced genomes where the imuB gene could be found. Only M.tuberculosis and M.bovis were included for simplification, although imuB can be found in several other Mycobacterial genomes. The names of the ORFs and species name abbreviations shown are the same adopted in the genome annotation of each organism. Abbreviations of organism names are indicated in the legend of Figure 2. The numbers indicate the bootstrap values >50%.
Characterization of the imuAB/dnaE2 expression
The genomic organization of the CC3213 (imuA), CC3212 (imuB) and CC3211 (dnaE2) genes suggests that they may constitute an operon (Figure 1). In fact, in P.putida, Sinorhizobium meliloti and Agrobacterium tumefaciens, the orthologs of these genes are co-expressed in a single RNA (17). In order to evaluate the co-expression of these genes in C.crescentus, we designed RT–PCR experiments with the oligonucleotides represented in Figure 4A. The results confirmed that the three genes are expressed as a single operon, and the polycistronic RNA can be detected both in UV-irradiated and non-irradiated exponentially growing cells.
Figure 4 Expression of the imuAB dnaE2 operon. (A) Schematic diagram for this operon indicating the position of the oligonucleotides (small arrows) used in the RT–PCR strategy to detect the co-expression of the three genes. Template RNA was extracted from exponentially growing cells, both from non-irradiated cultures and from cultures irradiated with 60 J/m2 of UV and recovered after 45 min. The RT–PCR products were visualized after electrophoresis in 1% agarose gel stained with ethidium bromide. (B) Expression of uvrA and imuA promoters in transcriptional fusions with the lacZ gene. ?-Galactosidase activity was measured 90 min after irradiation with UV light, both in the wild-type and recA strains.
In -proteobacteria, the LexA-binding site has been characterized for many species. The typical sequence GTTCN7GTTC is apparently well conserved in all species examined to date (34,35). The operon containing imuA, imuB and dnaE2 presents this typical LexA-binding sequence 66 bp upstream from the imuA start codon, indicating that this operon is SOS regulated. In order to examine whether the expression of this operon is regulated in response to DNA damage, we cloned the promoter of imuA upstream the reporter gene lacZ in the pRKlacZ290 vector. For comparison, the promoter of the uvrA gene, which also bears a putative -proteobacteria type SOS box 78 bp upstream the start codon, was cloned in the same vector. Determination of ?-galactosidase activity was performed both for the wild-type NA1000 strain, and for a derivative recA deletion mutant, constructed in this work. Both promoters were strongly induced 90 min after UV irradiation in a dose-responsive fashion in the wild-type strain (Figure 4B), although the expression of lacZ under control of the imuAB/dnaE2 promoter is much higher than the one observed for the uvrA promoter.
In our conditions, optimal promoter expression was achieved 90–120 min after irradiation (data not shown), which is in good agreement with the previous reports of Weigle reactivation time course experiments in this organism (19). On the other hand, damage-induced activity of both promoters is completely abolished in a recA-deleted strain, indicating that both are regulated in a RecA-dependent manner. These results confirm the prediction that both uvrA and the imuAB/dnaE2 operon are controlled by the SOS regulon in C.crescentus.
The increased activity of the imuA promoter resulting in the induction of all three genes in the operon was further investigated by quantitative RT–PCR experiments. In addition, the expression of the dinB gene (encoded by the CC2466 ORF), which does not present an SOS box in its putative promoter region, was also tested. The results shown in Figure 5 demonstrate that imuA, imuB and dnaE2 expression levels are increased similarly in response to UV damage. Furthermore, the damage induction of the three genes is abolished in the recA strain, indicating that they are coordinately upregulated by the SOS response. The SOS box-containing promoter upstream of imuA presumably mediates this coordinated expression. In contrast, the expression of dinB remains unaffected by DNA damage, suggesting that this gene is not involved in damage tolerance in C.crescentus.
Figure 5 Quantitative RT–PCR analysis of UV-induced gene expression. RNAs were extracted from exponentially growing recA and wild-type cells before and 45 min after irradiation with 60 J/m2 UVC. Relative levels of expression shown represent the average of two independent experiments done in triplicate.
Genetic characterization of damage-inducible mutagenesis in C.crescentus
Several strains disrupted in the imuA, imuB and dnaE2 genes, as well as in the dinB gene, were constructed (Table 1). These gene disruptions had no effect upon cell growth (data not shown). Compared with the wild type, all the single mutant strains deficient in the genes of the operon were slightly sensitive to UV (Figure 6A). In contrast, the dinB strain showed wild-type levels of UV resistance, indicating that DinB is not involved in UV damage tolerance. The double mutant imuB dnaE2 showed no increased sensitivity to UV, indicating that the products of these genes act in the same tolerance pathway. A dinB imuB double mutant was also constructed, to evaluate if DinB could be acting in DNA damage tolerance in an imuAB-deficient background. The dinB mutation had no effect on the resistance to UV, further indicating that this gene is not involved in UV damage resistance.
Figure 6 Effects of imuAB, dnaE2 and dinB mutations on UV resistance and mutagenesis in C.crescentus. (A) Survival curves. (B) Frequency of Rifr mutations induced by UV in the rpoB gene. The results shown are the mean of at least five independent experiments done in triplicate. Error bars indicate the SD.
The mutagenesis of these strains was determined by the rifampicin resistance assay. This assay detects point mutations in the rpoB gene, which encodes the ?-subunit of bacterial RNA polymerase, and is very useful due to the high conservation of the target gene, allowing its utilization in a wide variety of bacterial species (36,37). Figure 6B shows the mutagenesis induced by UV light in all the strains. It is clear that dinB depletion did not affect UV-induced mutagenesis, in contrast to the disruption of imuA, imuB and dnaE2. The UV-induced mutations are almost equally diminished in these three strains, and reduced to approximately the same extent in the imuB dnaE2 double mutant. Altogether, the results shown in Figure 6 demonstrate that the imuAB dnaE2 operon is involved in mutagenic repair of UV lesions, and suggest that the proteins encoded by this operon cooperate in the same pathway, as evidenced by the similar phenotypes of the single mutants, and by the epistatic effect of imuB and dnaE2 mutations. The dinB imuB strain presents the same levels of UV-induced mutagenesis as the imuB single mutant, confirming that DinB is not implicated in error-prone repair in the conditions tested.
Similar experiments were performed to investigate the effects of the UV mimetic drug mit C in C.crescentus. The effects of mit C on the survival of the different strains are much more pronounced than those of UV light (Figure 7A). The single mutants imuA, imuB and dnaE2 mutants are extremely sensitive to this condition, revealing that this operon plays a major role in the tolerance to the mit C-induced DNA lesions. The imuB dnaE2 double mutant does not exhibit an exacerbated phenotype, indicating again that these genes cooperate in the same pathway of damage tolerance. On the other hand, the dinB mutant is as resistant as the wild type, and the dinB imuB double mutant shows no further sensitivity enhancement, demonstrating that DinB is not involved in the tolerance to mit C-induced DNA lesions. Both the UV and the mit C sensitivity phenotypes of imuA could be complemented by vector pMR3213, containing the wild-type imuA allele. Similarly, the imuB strain could be complemented by cosmid 3A4, which contains integral imuA and imuB genes, but lacking dnaE2 (data not shown). These results confirm that the phenotypes of imuA and imuB strains are not due to negative effects on the expression of downstream genes in the operon.
Figure 7 Effects of imuAB, dnaE2 and dinB mutations on mit C resistance and mutagenesis in C.crescentus. (A) Survival curves. (B) Frequency of Rifr mutations induced by mit C in the rpoB gene. The results shown are the mean of at least four independent experiments done in triplicate. Error bars indicate the SD.
The mutagenesis induced by mit C was also investigated using the rifampicin assay (Figure 7B). The results show a dramatic decrease in mit C mutagenesis in all strains devoid of genes belonging to the damage-inducible operon of C.crescentus. On the other hand, dinB is clearly not required at all for mit C mutagenesis. These data confirm that the imuAB/dnaE2 operon accounts for most of the mutagenic DNA repair in C.crescentus cells exposed to UV light and mit C.
Mutational specificity of UV light in C.crescentus
The nature of imuAB/dnaE2-dependent and independent mutations caused by UV light was analyzed in the rpoB gene. Rifampicin-resistant mutants obtained after UV radiation of the wild-type, dnaE2 and imuB strains were randomly selected for sequencing of the rpoB cluster II of rifampicin resistance (37), which contained the Rifr mutations in all mutants analyzed. The mutation spectra observed are shown in Figure 8. The mutation signature of UV light is remarkably similar in dnaE2 and imuB strains. G:C A:T transitions are by far the most common type of substitution, constituting 90% of the base changes in both strains (Table 2). On the other hand, UV mutagenesis in the parental strain is significantly modified to a high proportion of G:C C:G transversions and tandem substitutions concentrated in a hotspot consisting of a run of three Cs, which gives rise to only G:C A:T mutations in dnaE2 and imuB strains (Figure 8). This increase in transversions and tandem substitutions in the parental strain is accompanied by concomitant reduction in the fraction of G:C A:T transitions, as shown in Table 2. These other types of mutations are probably an outcome of the mutagenic activity of the imuAB/dnaE2 operon of C.crescentus, since they are not observed in the dnaE2 and imuB mutants. G:C C:G transversions are not a significant part of the mutagenic spectrum of UV light in E.coli (37,38). These results indicate that imuB and dnaE2 cooperate in a lesion bypass pathway, which has a different specificity than that involving umuDC in E.coli.
Figure 8 UV mutation spectra in the rpoB gene. Nucleotides 1557–1625, comprising the cluster II of rifampicin resistance, are shown in this figure. All mutants analyzed had mutations in this region of the rpoB gene. Mutations in shaded background correspond to repetitions of the same base substitution, and tandem mutations are underlined. The total number of independent mutations sequenced is indicated for each strain.
Table 2 Frequencies of base substitutions in the rpoB gene induced by UV light
DISCUSSION
In the present work, we demonstrate that a highly conserved gene expression unit consisting of the genes imuAB and dnaE2, controlled by the SOS regulon, is responsible for most of the damage-inducible mutagenesis in C.crescentus. The genes imuA and imuB encode previously uncharacterized proteins, which are shown here for the first time to be involved in DNA damage tolerance. These genes are present in several genomes where umuDC orthologs cannot be found, such as the -proteobacteria branch, and it is intuitive to argue that they may functionally replace these proteins in damage-inducible mutagenesis in a wide variety of bacterial species. It is very interesting to note that, independently of being organized as single transcriptional units or as independent genes, imuB and dnaE2 always co-exist in bacterial genomes, strongly suggesting that their activities are interconnected.
The data obtained here clearly indicate that this operon plays a major role in damage-induced mutagenesis in C.crescentus, especially after mit C exposure. The lack of mit C mutagenesis in imuAB/dnaE2-deficient strains correlates well with the extreme sensitivity they show to this agent. Although mit C mutagenesis is mostly dependent on these genes, at least one-third of the UV-induced mutations are still generated in the imu- or dnaE2-deficient genetic backgrounds. These mutations are not caused by the action of the dinB gene, as evidenced by the phenotypes of dinB and imuB dinB mutant strains. The nature of this mutagenic process remains to be elucidated. In E.coli, it has been shown that umu-independent UV mutagenesis may occur, depending on the experimental system used (39). Also, in B.subtilis, residual UV mutagenesis can be detected in varying levels, even in cells devoid of both Y-family DNA polymerases (40,41). It is interesting to note that no other protein belonging to the UmuC superfamily can be found in the C.crescentus genome, and neither there are polB orthologs. Thus, imuAB/dnaE2-independent UV mutagenesis might occur via a still uncharacterized pathway. Recently, it has been suggested that the dnaE gene of C.crescentus may also be a part of the SOS regulon (42). Overexpression of the replicative polymerase encoded by dnaE may account for the imuAB/dnaE2-independent UV mutagenesis. On the other hand, mit C mutagenesis shows an absolute dependence on the activity of the imuAB dnaE2 operon, indicating that DnaE and the ImuAB DnaE2 machinery may have different lesion bypass capacities.
The mutation spectra of UV-induced lesions in C.crescentus show a dramatic qualitative difference between wild-type and imuB or dnaE2 strains. G:C A:T transitions, a hallmark of UV mutagenesis in E.coli, are observed in the mutant strains, while wild-type cells also have a high proportion of tandem substitutions and G:C C:G transversions. This novel mutational signature, dependent on the activity of the imuAB/dnaE2 operon, probably reflects a difference in polymerase preference for nucleotide incorporation during TLS. On the other hand, the mutations induced by UV light in M.tuberculosis in a dnaE2-dependent manner are different from the observed in C.crescentus, resembling the UV mutagenic spectrum of E.coli (14). More detailed studies of both systems are still needed to clarify this discrepancy, although it is clear that there are fundamental differences between these organisms, since imuA orthologs are not present in M.tuberculosis and related bacteria. In addition, the use of the rpoB gene as target for mutational analysis has intrinsic limitations, such as the underestimation of frameshifts and the limited number of mutations detectable as gain of function alleles for Rifr. The development of more sophisticated genetic tools for the study of mutagenesis is mandatory for a deeper understanding of the basic mechanisms of this aspect of DNA metabolism in C.crescentus.
The DNA polymerase III holoenzyme is responsible for chromosomal replication. The -subunit of DNA polymerase III, encoded by the dnaE gene, is essential in E.coli (43). In C.crescentus, the involvement of the dnaE gene (ORF CC1926) in DNA replication has also been demonstrated previously (31). However, in M.tuberculosis, the deletion of dnaE2 causes no growth defect (14), similar to what was observed for C.crescentus. It was demonstrated years ago that dnaE is required for UV mutagenesis in E.coli (44,45). Currently, it is thought that successful TLS takes place after a blockage of Pol III and formation of a RecA filament on single-stranded DNA. Then, Y-family polymerases displace the Pol III catalytic subunit and mediate lesion bypass, synthesizing patches of DNA sufficiently sized to allow the hand-off of replication back to Pol III. These polymerase switches are mediated by interactions with the ?-clamp, the processivity factor of Pol III (46–48). The observation that several bacterial species bear a second copy of the dnaE gene, which seems to be solely devoted to DNA damage response, is very intriguing. The presence of this second, damage inducible, Pol III catalytic subunit may be an alternative manner some bacteria have found to control polymerase switch during lesion bypass. Additionally, DnaE2 may itself act as a TLS polymerase, as shown in B.subtilis and S.pyogenes (12,13).
In spite of the functional and phylogenetic characterization provided in this work, the actual biochemical functions of imuA and imuB genes remain to be elucidated. However, the genetic data shown here indicate that these genes cooperate with dnaE2 in an error-prone repair pathway. ImuB is a member of the UmuC superfamily, and it is tempting to speculate about a model in which ImuB and DnaE2 would act in concert, as shown for Y-family polymerases and Pol III in TLS in E.coli. This hypothesis is consistent with the genetic requirement for both imuB and dnaE2 in damage-inducible mutagenesis shown in this work, and is further supported by the epistatic effect of these gene disruptions. This model would present a functional conservation of the overall TLS scenario in bacteria, although natural selection has chosen different role-players within different domains of the bacterial kingdom. Shedding light on the reasons why different bacteria should have distinct mutagenic repair machinery is a challenging question, which deserves further examination.
SUPPLEMENTARY MATERIAL
Supplementary Material is available at NAR Online.
ACKNOWLEDGEMENTS
The authors would like to thank Dr Craig Stephens (Santa Clara University) for the gift of the 3A4 cosmid. Financial support was obtained from FAPESP (S?o Paulo, Brazil) and CNPq (Brasília, Brazil). R.S.G. is a Postdoctoral fellow from FAPESP. Funding to pay the Open Access publication charges for this article was provided by FAPESP.
REFERENCES
Rangarajan, S., Woodgate, R., Goodman, M.F. (1999) A phenotype for enigmatic DNA polymerase II: a pivotal role for pol II in replication restart in UV-irradiated Escherichia coli Proc. Natl Acad. Sci. USA, 96, 9224–9229 .
Napolitano, R., Janel-Bintz, R., Wagner, J., Fuchs, R.P. (2000) All three SOS-inducible DNA polymerases (Pol II, Pol IV and Pol V) are involved in induced mutagenesis EMBO J., 19, 6259–6265 .
Wagner, J., Gruz, P., Kim, S.R., Yamada, M., Matsui, K., Fuchs, R.P., Nohmi, T. (1999) The dinB gene encodes a novel E.coli DNA polymerase, DNA pol IV, involved in mutagenesis Mol. Cell, 4, 281–286 .
Ohmori, H., Friedberg, E., Fuchs, R.P., Goodman, M.F., Hanaoka, F., Hinkle, D., Kunkel, T.A., Lawrence, C.W., Livneh, Z., Nohmi, T. (2001) The Y-family of DNA polymerases Mol. Cell, 8, 7–8 .
Kim, S.R., Maenhaut-Michel, G., Yamada, M., Yamamoto, Y., Matsui, K., Sofuni, T., Nohmi, T., Ohmori, H. (1997) Multiple pathways for SOS-induced mutagenesis in Escherichia coli: an overexpression of dinB/dinP results in strongly enhancing mutagenesis in the absence of any exogenous treatment to damage DNA Proc. Natl Acad. Sci. USA, 94, 13792–13797 .
Hersh, M.N., Ponder, R.G., Hastings, P.J., Rosenberg, S.M. (2004) Adaptive mutation and amplification in Escherichia coli: two pathways of genome adaptation under stress Res. Microbiol., 155, 352–359 .
Elledge, S.J. and Walker, G.C. (1983) Proteins required for ultraviolet light and chemical mutagenesis: identification of the products of the umuC locus of Escherichia coli J. Mol. Biol., 164, 175–192 .
Shinagawa, H., Kato, T., Ise, T., Makino, K., Nakata, A. (1983) Cloning and characterization of the umu operon responsible for inducible mutagenesis in Escherichia coli Gene, 23, 167–174 .
Reuven, N.B., Arad, G., Maor-Shoshani, A., Livneh, Z. (1999) The mutagenesis protein UmuC is a DNA polymerase activated by UmuD', RecA, and SSB and is specialized for translesion replication J. Biol. Chem., 274, 31763–31766 .
Tang, M., Shen, X., Frank, E.G., O'Donnell, M., Woodgate, R., Goodman, M.F. (1999) UmuD'(2)C is an error-prone DNA polymerase, Escherichia coli pol V Proc. Natl Acad. Sci. USA, 96, 8919–8924 .
Tippin, B., Pham, P., Goodman, M.F. (2004) Error-prone replication for better or worse Trends Microbiol., 12, 288–295 .
Bruck, I., Goodman, M.F., O'Donnell, M. (2003) The essential C family DnaE polymerase is error-prone and efficient at lesion bypass J. Biol. Chem., 278, 44361–44368 .
Le Chatelier, E., Bécherel, O.J., d'Alen?on, E., Canceill, D., Ehrlich, S.D., Fuchs, R.P., Jannière, L. (2004) Involvement of DnaE, the second replicative DNA polymerase from Bacillus subtilis, in DNA mutagenesis J. Biol. Chem., 279, 1757–1767 .
Boshoff, H.I., Reed, M.B., Barry, C.E., III, Mizrahi, V. (2003) DnaE2 polymerase contributes to in vivo survival and the emergence of drug resistance in Mycobacterium tuberculosis Cell, 113, 183–193 .
Davis, E.O., Dullaghan, E.M., Rand, L. (2002) Definition of the mycobacterial SOS box and use to identify LexA-regulated genes in Mycobacterium tuberculosis J. Bacteriol., 184, 3287–3295 .
Rand, L., Hinds, J., Springer, B., Sander, P., Buxton, R.S., Davis, E.O. (2003) The majority of inducible DNA repair genes in Mycobacterium tuberculosis are induced independently of RecA Mol. Microbiol., 50, 1031–1042 .
Abella, M., Erill, I., Jara, M., Mazon, G., Campoy, S., Barbe, J. (2004) Widespread distribution of a lexA-regulated DNA damage-inducible multiple gene cassette in the Proteobacteria phylum Mol. Microbiol., 54, 212–222 .
Nierman, W.C., Feldblyum, T.V., Laub, M.T., Paulsen, I.T., Nelson, K.E., Eisen, J.A., Heidelberg, J.F., Alley, M.R., Ohta, N., Maddock, J.R., et al. (2001) Complete genome sequence of Caulobacter crescentus Proc. Natl Acad. Sci. USA, 98, 4136–4141 .
Bender, R.A. (1984) Ultraviolet mutagenesis and inducible DNA repair in Caulobacter crescentus Mol. Gen. Genet., 197, 399–402 .
Ely, B. (1991) Genetics of Caulobacter crescentus Methods Enzymol., 204, 372–384 .
Prentki, P. and Krisch, H.M. (1984) In vitro insertional mutagenesis with a selectable DNA fragment Gene, 29, 303–313 .
Miller, J.H. A Short Course in Bacterial Genetics—A Laboratory Manual and Handbook for Escherichia Coli and Related Bacteria, (1992) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press pp. 72–74 .
Pfaffl, M.W. (2001) A new mathematical model for relative quantification in real-time RT–PCR Nucleic Acids Res., 29, e45 .
Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res., 25, 3389–3402 .
Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G. (1997) The ClustalX Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools Nucleic Acids Res., 25, 4876–4882 .
Nicholas, K.B. and Nicholas, H.B., Jr. (1997) GeneDoc: a tool for editing and annotating multiple sequence alignments .
Felsenstein, J. (1989) PHYLIP Phylogeny Inference Package (Version 3.2) Cladistics, 5, 164–166 .
Saitou, N. and Nei, M. (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees Mol. Biol. Evol., 4, 406–425 .
Page, R.D.M. (1996) TREEVIEW: an application to display phylogenetic trees on personal computers Comput. Appl. Biosci., 12, 357–359 .
Martins-Pinheiro, M., Galhardo, R.S., Lage, C., Lima-Bessa, K.M., Aires, K.A., Menck, C.F. (2004) Different patterns of evolution for duplicated DNA repair genes in bacteria of the Xanthomonadales group BMC Evol. Biol., 4, 29 .
Lo, T., van Der Schalie, E., Werner, T., Brun, Y.V., Din, N. (2004) A temperature-sensitive mutation in the dnaE gene of Caulobacter crescentus that prevents initiation of DNA replication but not ongoing elongation of DNA J. Bacteriol., 186, 1205–1212 .
Weinel, C., Nelson, K.E., Tümmier, B. (2002) Global features of the Pseudomonas putida KT2440 genome sequence Environ. Microbiol., 4, 809–818 .
Bruggemann, H., Henne, A., Hoster, F., Liesegang, H., Wiezer, A., Strittmatter, A., Hujer, S., Durre, P., Gottschalk, G. (2004) The complete genome sequence of Propionibacterium acnes, a commensal of human skin Science, 305, 671–673 .
Fernandez de Henestrosa, A.R., Rivera, E., Tapias, A., Barbe, J. (1998) Identification of the Rhodobacter sphaeroides SOS box Mol. Microbiol., 28, 991–1003 .
Tapias, A. and Barbe, J. (1999) Regulation of divergent transcription from the uvrA-ssb promoters in Sinorhizobium meliloti Mol. Gen. Genet., 262, 121–130 .
Kim, M., Wolff, E., Huang, T., Garibyan, L., Earl, A.M., Battista, J.R., Miller, J.H. (2004) Developing a genetic system in Deinococcus radiodurans for analyzing mutations Genetics, 166, 661–668 .
Garibyan, L., Huang, T., Kim, M., Wolff, E., Nguyen, A., Nguyen, T., Diep, A., Hu, K., Iverson, A., Yang, H., et al. (2003) Use of the rpoB gene to determine the specificity of base substitution mutations on the Escherichia coli chromosome DNA Repair (Amst.), 2, 593–608 .
Livneh, Z., Cohen-Fix, O., Skaliter, R., Elizur, T. (1993) Replication of damaged DNA and the molecular mechanism of ultraviolet light mutagenesis Crit. Rev. Biochem. Mol. Biol., 28, 465–513 .
Christensen, J.R., LeClerc, J.E., Tata, P.V., Christensen, R.B., Lawrence, C.W. (1988) UmuC function is not essential for the production of all targeted lacI mutations induced by ultraviolet light J. Mol. Biol., 203, 635–641 .
Sung, H.M., Yeamans, G., Ross, C.A., Yasbin, R.E. (2003) Roles of YqjH and YqjW, homologs of the Escherichia coli UmuC/DinB or Y superfamily of DNA polymerases, in stationary-phase mutagenesis and UV-induced mutagenesis of Bacillus subtilis J. Bacteriol., 185, 2153–2160 .
Duigou, S., Ehrlich, S.D., Noirot, P., Noirot-Gros, M.F. (2004) Distinctive genetic features exhibited by the Y-family DNA polymerases in Bacillus subtilis Mol. Microbiol., 54, 439–451 .
Erill, I., Jará, M., Salvador, N., Escribano, M., Campoy, S., Barbe, J. (2004) Differences in LexA regulon structure among Proteobacteria through in vivo assisted comparative genomics Nucleic Acids Res., 32, 6617–6626 .
McHenry, C.S. (1998) DNA polymerase III holoenzyme of Escherichia coli Annu. Rev. Biochem, 57, 519–550 .
Bridges, B.A. and Bates, H. (1990) Mutagenic DNA repair in Escherichia coli. XVIII. Involvement of DNA polymerase III alpha-subunit (DnaE protein) in mutagenesis after exposure to UV light Mutagenesis, 5, 35–38 .
Bridges, B.A. and Mottershead, R.P. (1976) Mutagenic DNA repair in Escherichia coli. III. Requirement for a function of DNA polymerase III in ultraviolet-light mutagenesis Mol. Gen. Genet., 144, 53–58 .
Sutton, M.D., Opperman, T., Walker, G.C. (1999) The Escherichia coli SOS mutagenesis proteins UmuD and UmuD' interact physically with the replicative DNA polymerase Proc. Natl Acad. Sci. USA, 96, 12373–12378 .
Bunting, K.A., Roe, S.M., Pearl, L.H. (2003) Structural basis for recruitment of translesion DNA polymerase PolIV/DinB to the beta-clamp EMBO J., 22, 5883–5892 .
Fujii, S. and Fuchs, R.P. (2004) Defining the position of switches between replicative and bypass DNA polymerases EMBO J., 23, 4342–4352 .
Evinger, M. and Agabian, N. (1977) Envelope-associated nucleoid from Caulobacter crescentus stalked and swarmer cells J. Bacteriol., 132, 294–301 .
Simon, R., Prieffer, U., Puhler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria Biotechnology, 1, 784–790 .
Tsai, J.W. and Alley, M.R. (2000) Proteolysis of the McpA chemoreceptor does not require the Caulobacter major chemotaxis operon J. Bacteriol., 182, 504–507 .
Roberts, R.C., Toochinda, C., Avedissian, M., Baldini, R.L., Gomes, S.L., Shapiro, L. (1996) Identification of a Caulobacter crescentus operon encoding hrcA, involved in negatively regulating heat-inducible transcription and the chaperone gene grpE J. Bacteriol., 178, 1829–1841 .
Gober, J.W. and Shapiro, L. (1992) A developmentally regulated Caulobacter flagellar promoter is activated by 3' enhancer and IHF binding elements Mol. Biol. Cell, 3, 913–916 .
Alley, M.R., Gomes, S.L., Alexander, W., Shapiro, L. (1991) Genetic analysis of a temporally transcribed chemotaxis gene cluster in Caulobacter crescentus Genetics, 129, 333–341 .(Rodrigo S. Galhardo, Raquel P. Rocha, Ma)
*To whom correspondence should be addressed at Department of Microbiology, Institute of Biomedical Sciences, University of S?o Paulo, Avenue Professor Lineu Prestes, 1374, S?o Paulo, SP 05508-900, Brazil. Tel: +55 11 3091 7499; Fax: +55 11 3091 7354; Email: cfmmenck@usp.br
ABSTRACT
DNA polymerases of the Y-family, such as Escherichia coli UmuC and DinB, are specialized enzymes induced by the SOS response, which bypass lesions allowing the continuation of DNA replication. umuDC orthologs are absent in Caulobacter crescentus and other bacteria, raising the question about the existence of SOS mutagenesis in these organisms. Here, we report that the C.crescentus dinB ortholog is not involved in damage-induced mutagenesis. However, an operon composed of two hypothetical genes and dnaE2, encoding a second copy of the catalytic subunit of Pol III, is damage inducible in a recA-dependent manner, and is responsible for most ultraviolet (UV) and mitomycin C-induced mutations in C.crescentus. The results demonstrate that the three genes are required for the error-prone processing of DNA lesions. The two hypothetical genes were named imuA and imuB, after inducible mutagenesis. ImuB is similar to proteins of the Y-family of polymerases, and possibly cooperates with DnaE2 in lesion bypass. The mutations arising as a consequence of the activity of the imuAB dnaE2 operon are rather unusual for UV irradiation, including G:C to C:G transversions.
INTRODUCTION
DNA damage-induced mutagenesis is, in a major extent, an active process that requires specialized DNA polymerases able to perform translesion synthesis (TLS). In Escherichia coli, three SOS-regulated DNA polymerases are involved in damage tolerance: Pol II, encoded by the polB gene, Pol IV and Pol V. The major role of Pol II is the rescue of stalled replication forks (1), although it can bypass some types of DNA damage (2). dinB (also called dinP) encodes a member of the Y-family of DNA polymerases, DNA Pol IV (3,4). This enzyme is responsible for untargeted SOS mutagenesis of phage , and induces frameshift mutations when overexpressed (5). More recently, dinB has been associated with the phenomenon of mutation under stress conditions, or adaptive mutagenesis .
In E.coli, ultraviolet (UV) and chemical mutagenesis is heavily dependent on the SOS-regulated umuDC operon (7,8). The UmuC protein has been identified as a DNA polymerase (Pol V), a distributive enzyme able to perform TLS and highly inaccurate even when using undamaged templates (9,10), which interacts with the processed form of the co-expressed protein UmuD. Both in vivo and in vitro experiments have shown that efficient TLS requires the complex, RecA, Ssb, the catalytic subunit of DNA Pol III encoded by the dnaE gene, and the ? clamp .
DNA polymerase III is a multiprotein complex responsible for the replication of the bacterial chromosome. Although the Y-family DNA polymerases, specialized in TLS, seem to be ubiquitous throughout nature, it has been shown that the subunit of Pol III, which bears the catalytic polymerase activity, is also able to perform TLS. In Bacillus subtilis, the dnaE gene is SOS regulated and the purified enzyme can perform TLS in vitro, an activity also shown for the Streptococcus pyogenes protein (12,13). Depletion of B.subtilis DnaE inhibits UV-induced mutagenesis, although it is difficult to ascertain whether this requirement reflects the need for this enzyme to extend the bypass reaction performed by Y-family DNA polymerases, or results from actual lack of lesion bypass (13).
It has recently been shown that a second copy of the dnaE gene mediates SOS mutagenesis in Mycobacterium tuberculosis (14). The expression of this gene is regulated by the SOS response in this organism, being induced by DNA damaging agents in a recA-dependent manner (15,16). The dnaE gene is presented in duplicate in several bacterial genomes, most frequently in a potential operon with two hypothetical genes, one of which is a dinB-related one. This widespread operon has been shown to be regulated by LexA in Pseudomonas putida, indicating that it is involved in DNA damage response (17). This operon is present in several -proteobacteria genomes completely sequenced to date, including C.crescentus (18).
In C.crescentus, a recA-dependent inducible repair system was demonstrated through the existence of Weigle reactivation of irradiated phage (19). This organism is also mutable by UV radiation, suggesting the existence of error-prone pathways, although umuDC orthologs are absent from its genome (18,19). In this work, we investigated the involvement of the dnaE2-containing gene cassette and the dinB ortholog in damage-induced mutagenesis, detected in rifampicin resistance assays. The results show that the dnaE2 operon is upregulated in response to DNA damage in a recA-dependent manner, and demonstrate that all three genes that compose this expression unit are involved in UV and mitomycin C-induced mutagenesis. On the other hand, dinB is not necessary for mutagenic repair. We propose to name the genes co-expressed with dnaE2, after inducible mutagenesis, as imuA and imuB, and show that the imuAB/dnaE2-dependent mutational spectra of UV-induced lesions is different from the prototypical E.coli one, including a high proportion of G:C C:G transversions.
MATERIALS AND METHODS
Bacterial strains and plasmids
Bacterial strains and plasmids used in this study are shown in Table 1. C.crescentus strains were grown in PYE medium (20) at 30°C with constant shaking. Plasmids were introduced into C.crescentus by conjugation with E.coli S17-1 strain. When appropriated, the culture medium was supplemented with kanamycin (50 μg/ml), nalidixic acid (25 μg/ml), spectinomycin (50 μg/ml) or tetracycline (1 μg/ml). E.coli strain DH10B (Invitrogen, CA) was used for cloning purposes. E.coli was grown at 37°C in Luria–Bertani medium supplemented with ampicillin (100 μg/ml), kanamycin (50 μg/ml) or tetracycline (15 μg/ml) when necessary.
Table 1 Bacterial strains and plasmids used in this study
Genetic disruptions were achieved using the pNPTS138 vector, and selecting for two consecutive recombination events. First, Kanr conjugants of C.crescentus were selected in the screening for the first recombination event. Loss of the plasmid after the second recombination was selected in PYE media containing 3% sucrose. When appropriated, a concomitant Spcr selection was also applied. Strains generated in this way were analyzed by Southern blot to confirm gene disruptions. The construction of gene targeting plasmids was performed with PCR products amplified with the primers shown in Supplementary Table 1S. These products were cloned in the pGEM-T easy vector and sequenced to assure sequence integrity. For disruption of recA, fragments amplified with primers recaA x recaB, and recaC x recaD were fused in the pNPTS138 vector using the restriction sites introduced, generating a 1.8 kb fragment containing the first 81 bp fused to the last 75 bp of recA plus flanking regions, leading to an in-frame deletion of recA. In-frame deletions were constructed in the same way for imuA and imuB using, respectively, primers CC3213A to D and CC3212A to D. imuA disruption resulted in a deletion containing only the first 36 and the last 54 bp of the gene, and imuB disruption resulted in a deletion containing only the first 63 and the last 81 bp of the gene. dnaE2 disruption was performed by the insertion of the specr cassette (21) in the natural EcoRI site of a 1629 bp fragment of this gene, amplified with the dnaE2re and dnaE2fo primers. The insertion in the EcoRI site resulted in a predicted truncation at position 598 of the protein, leading to loss of the last 485 amino acids. dinB knock-out was constructed by insertion of the specr cassette in the EcoRV site of a 1.2 kb fragment, amplified with primers dinbfo and dinbre. This insertion resulted in a predicted truncation at position 167 of the 419 amino acid DinB protein.
PCRs were also carried out to amplify promoter regions of uvrA and imuA genes. The primers puvrafo and puvrare were used to obtain a fragment ranging from –492 to +54 of the start codon of uvrA . Primers p3213fo and p3213re amplified a fragment ranging from –513 to +74 of the imuA start codon. Both fragments were cloned in the pRKlacZ290 vector using the restriction sites introduced in the primers.
In order to promote complementation of the phenotypes of the imuA strain, a full-length gene including the promoter region was amplified with the 3213A and 3213D oligonucleotides, and cloned in the low copy number pMR20 vector, using the restriction sites introduced in the primers.
Survival and mutagenesis experiments
Caulobacter crescentus strains were grown until mid-log phase (5 x 108 cells/ml), and then submitted to the different treatments. A 5 ml batch of culture was UV-irradiated with a Philips TUV 15W/G15T8 germicidal lamp (mainly 254 nm) in Petri dishes. Aliquots were removed before and after irradiation for serial dilutions and plating in PYE medium, for the determination of viable cell counts after 48 h incubation at 30°C. For mutagenesis experiments, a 200 μl aliquot of irradiated cells was inoculated in 1 ml of PYE liquid medium and cultivated overnight, to allow mutation fixation. Then, the cultures were submitted to serial dilutions and plating on PYE medium for viable cell counts. The cultures were also plated in PYE medium containing 100 μg/ml rifampicin to score Rifr cells. Mutation frequencies were calculated by dividing the numbers of mutants by the estimated total number of cells analyzed in the experiments. Mitomycin C (mit C) survival and mutagenesis experiments were conducted in a similar manner, with drug treatments performed in 2 ml of culture for 1 h at 30°C. For mit C mutagenesis, cells were centrifuged and resuspended in PYE before inoculating for overnight incubation, to remove the drug from the medium.
Determination of the mutation spectra in the rpoB gene
Rifampicin-resistant clones, obtained after UV irradiation in the experiments described above, were cultivated overnight for the extraction of chromosomal DNA. The primers rpoB2fo and rpoB2re (Supplementary Table 1S) were used in PCRs designed for the amplification of the region II of rifampicin resistance. PCR products were purified and sequenced in both strands using rpoB2fo and rpoB2re primers with the Big Dye Terminator Cycle Sequencing kit and analyzed in the ABI Prism 3100 Genetic Analyzer (Applied Biosystems, CA). Only independent mutations from each experiment were scored.
RT–PCR experiments and ?-galactosidase assays
The operon organization of genes CC3211, CC3212 and CC3213 was assayed through RT–PCR experiments. The primers 1112A and 1112B were used to check imuB/dnaE2 co-expression, and primers 1213A and 1213B were used to check imuA/imuB co-expression. RNA from exponentially growing cells was extracted with Trizol? reagent (Invitrogen), and treated with DNAseI to eliminate contaminant DNA. RT–PCR experiments were carried out with 300 ng of RNA using the Superscript One Step kit (Invitrogen), in the following cycling conditions: 50°C for 20 min; 94°C for 2 min; 40 cycles of 95°C for 1 min; 61°C for 30 s; 72°C for 50 s; followed by incubation at 72°C for 7 min and cooling to 4°C. A control for DNA contamination of samples was performed with PCR lacking reverse transcriptase.
Measurements of promoter activity with lacZ transcriptional fusions were performed after UV irradiation, in the same conditions used for survival experiments (see above), allowing cells to recover in PYE medium for 90 min after treatment. After recovery, cells were assayed for ?-galactosidase activity as described previously (22).
Real-time analysis of gene expression
Relative expression of imuA, imuB, dnaE2 and dinB after UV irradiation was determined with quantitative PCR experiments. An aliquot of 2 μg of RNA pre-treated with DNAse I were used as template for total cDNA synthesis in 20 μl reactions with random hexamers using the Superscript First-Strand Synthesis System for RT–PCR (Invitrogen). For real-time PCR, an amount of cDNA corresponding to 25 ng of input RNA was used in each reaction. Reactions were performed with the SYBR Green PCR Master Mix (Applied Biosystems), and analyzed in the ABI 7500 Real-Time System. Relative expression levels were calculated as described previously (23), using the rpoD gene as the endogenous control. Primers used in this analysis are listed in Supplementary Table 1S.
Phylogenetic analyses
Protein sequences of genes CC3211, CC3212 and CC3213 were obtained from the National Center of Biotechnology Information GenBank database (www.ncbi.nlm.nih.gov/Entrez/). The accession numbers are respectively NP_422005 , NP_422006 and NP_422007 . Orthologous protein sequences from complete genomes were obtained by BLAST analyses (24). The lack of orthologs in several genomes was ascertained by exhaustive BLAST analysis.
For phylogenetic analysis, the protein sequences were aligned using the ClustalX multiple sequence alignment program (25) with manual adjustment using Genedoc v2.6.001 (26) (www.psc.edu/biomed/genedoc). Only unambiguously aligned positions (excluding poorly conserved and gap regions) were used. Phylogenetic trees were generated for each group of protein homologs from sequence alignments using the PHYLIP program version 3.5 (27). Distance analyses were performed using the neighbor-joining method in PHYLIP, with the distance PAM matrix model (28). Bootstrap support (resampled 1000 times) was calculated, and strict consensus trees constructed. Only bootstrap values >50% are shown. The consensus trees obtained were viewed through TreeView software (29). The same set of prokaryotic species was used in all analyses, and includes CC3211 (dnaE2) and CC3212 (imuB) orthologs from fully sequenced genomes and E.coli proteins.
RESULTS
Genomic context and phylogenetic analyses
The CC3211 gene of C.crescentus encodes a second copy of the dnaE gene, hereafter referred to as dnaE2. dnaE2 is in a conserved organization with two genes encoding hypothetical proteins, CC3212 (imuB) and CC3213 (imuA), similar to other proteobacteria (Figure 1). This operon organization is eventually associated with a duplication of lexA, such as in P.putida. However, this is not the case for C.crescentus and related bacteria, which have only one copy the lexA gene (30). It was previously shown that the operon organization of imuAB and dnaE2 is conserved in all genomes of proteobacteria containing these genes (17). However, other bacterial species, such as several Actinomycetales and the planctomycete Rhodopirellula baltica, also possess a dnaE2 gene, although in different genomic contexts (Figure 1). A total of six different gene arrangements were found in bacterial genomes containing dnaE2. Remarkably, the large majority of organisms presenting dnaE2 also possess an imuB ortholog, although not necessarily in an operon organization. In R.baltica, imuA and imuB form a potential operon, which does not include dnaE2, located apart in the genome. The M.tuberculosis imuB gene, encoded by the Rv3394c ORF, is in an operon with Rv3395c, a putative RecA-like protein. Most interestingly, both Rv3395c and Rv3394c are damage inducible in M.tuberculosis, and the promoter of Rv3395c has been shown to contain a Mycobacterial SOS box (14,16). A similar picture can be observed in the genomes of other Actinomycetales, except Propionibacterium acnes, where the Rv3395c ortholog forms a putative operon with both imuB and dnaE2. Although proteins with significant similarity to ImuA cannot be found in Actinomycetales, the genomic context analysis suggests that the RecA-like protein encoded by M.tuberculosis Rv3395 may represent a functional counterpart of ImuA in these organisms. It is evident that several rearrangements of the genomic organization of dnaE2 and imuB have taken place during evolution, but notably these genes always co-exist in bacterial genomes. These data indicate that the imuB and dnaE2 genes compose a highly conserved co-evolving genetic system, which is likely to be involved in DNA damage response.
Figure 1 Prototypical genomic organizations of dnaE2 and accompanying imuA and imuB genes. The same organization found in C.crescentus is also present in -proteobacterial genomes, Pseudomonas aeruginosa, Vibrio vulnificus, Vibrio parahaemolyticus, Ralstonia solanacearum, Bordetella bronchiseptica and Bordetella parapertussis. A duplication of lexA preceding the operon, as found in P.putida, is also present in Methylococcus capsulatus, Pseudomonas syringae, Xanthomonas campestris and Xanthomonas axonopodis. The M.tuberculosis gene arrangement is conserved in all species of the genera Mycobacterium and Corynebacterium, and in Nocardia farcinica. Small boxes represent known SOS operators, and interrupted lines indicate that the genes are not contiguous in the genome. Loci numbers as provided by the genome annotation are indicated above, the arrows representing genes, and our annotation is given inside the arrows. Different original gene annotations are indicated in parentheses.
Phylogenetic analyses of the DnaE proteins of the microorganisms presenting such duplication indicate that this is most probably a very ancestral duplication event, present in the vast majority of fully sequenced genomes from -proteobacteria, but also found in other proteobacteria and in Actinomycetales (Figure 2). The branch of DnaE includes the E.coli protein and the product of the CC1926 gene, which seems to be involved in DNA replication in C.crescentus (31). DnaE2 orthologs clearly form a distinct branch, including the DnaE2 from C.crescentus. The prototypical member of this branch is the M.tuberculosis protein Rv3370c. This protein has been found to be damage inducible, and implicated in cell survival and mutagenesis after UV treatment (14).
Figure 2 Phylogenetic relationships between DnaE and DnaE2. Phylogenetic analyses included DnaE and DnaE2 proteins of all fully sequenced genomes where such duplication could be found. Only M.tuberculosis and Mycobacterium bovis were included for simplification, although the duplication can be found in several other Mycobacterial genomes. The names of the ORFs and species name abbreviations shown are the same adopted in the genome annotation of each organism. Abbreviations of organism names are as follows: AGR, A.tumefaciens str. C58; b, E.coli K12; BB, B.bronchiseptica RB50; bll, Bradyrhizobium japonicum USDA 110; BMA, Burkholderia mallei ATCC 23344; BMEI, Brucella melitensis 16M; BPP, B.parapertussis 12822; BPSL, Burkholderia pseudomallei K96243; BR, Brucella suis 1330; CC, C.crescentus CB15; CE, Corynebacterium efficiens YS-314; DIP, Corynebacterium diphtheriae NCTC 13129; Mb, M.bovis AF2122/97; MCA, M.capsulatus str. Bath; mll and mlr, Mesorhizobium loti MAFF303099; NCgl, Corynebacterium glutamicum ATCC 13032; Nfa, N.farcinica IFM 10152; PA, Pseudomonas aeruginosa PAO1; PP, P.putida KT2440; PPA, P.acnes KPA171202; PSPTO, P.syringae pv. tomato str. DC3000; RB, R.baltica SH 1; RPA, Rhodopseudomonas palustris CGA009; Rsc and Rsp, R.solanacearum GMI1000; Rv, M.tuberculosis H37Rv; SAV, S.avermitilis MA-4680; Smc and Sma, S.meliloti 1021; VP, V.parahaemolyticus RIMD 2210633; VV, V.vulnificus YJ016; XAC, X.axonopodis pv. citri str. 306; and XCC, Xanthomonas campestris pv. campestris str. ATCC 33913. The numbers indicate the bootstrap values >50%.
The CC3213 ortholog from P.putida, PP3117, was annotated as sulA2 (32). However, PSI-BLAST analyses of the protein encoded by CC3213 shows only very low levels of similarity to SulA (24), and also some similarity to RecA/RadA recombinases (data not shown). A phylogenetic tree of this protein and its orthologs has been presented previously (17), but the levels of protein similarities between CC3213 orthologs and known proteins, such as SulA and RecA, are too low to precisely ascribe any protein function. We tested the filamentaion in both the wild type and the derivative CC3213 deletion strain in several conditions by microscopic examination. C.crescentus shows a filamentous morphology after prolonged exposure to mit C, and the deletion of CC3213 did not affect this cellular response, indicating that this gene is not involved in cell division suppression (Supplementary Figure 1S). Therefore, we propose to name the CC3213 gene imuA, after inducible mutagenesis, given the functional characterization conducted (see below).
Similarly, CC3212 orthologs identified by BLAST analysis present diverse annotations: the P.putida gene PP3118 was annotated as dinP (32), and the PPA1651 gene of P.acnes was annotated as umuC (33), while most other orthologs (including CC3212 itself) were annotated as genes with unknown function. In this case, phylogenetic analysis was conducted including all fully sequenced genomes where CC3212 orthologs were found, for comparison between these proteins and DinB, the closest related protein of known function. All organisms with a CC3212 ortholog also have one or more typical dinB genes. The product of CC3212 and its orthologs form a branch that is distinct from the DinB proteins of the same organisms, indicating that this is a divergent DinB/UmuC homolog (Figure 3). The only possible exception is the SAV6556 protein of Streptomyces avermitilis, which cannot be surely assigned as ImuB through phylogenetic analyses, although it is present in a putative operon with dnaE2 (Figure 1). Protein similarity comparison among these proteins reinforces the difference between ImuB and DinB, e.g. E.coli DinB and UmuC proteins share 45% identity, while C.crescentus DinB and the ImuB proteins share only 35% identity. Therefore, we concluded that the CC3212 gene product is a member of a distinct branch in the UmuC superfamily of proteins, and not a DinB ortholog. We named this gene imuB, in accordance with the nomenclature adopted for CC3213.
Figure 3 Phylogenetic relationships between ImuB and DinB. Phylogenetic analyses included ImuB and DinB proteins of all fully sequenced genomes where the imuB gene could be found. Only M.tuberculosis and M.bovis were included for simplification, although imuB can be found in several other Mycobacterial genomes. The names of the ORFs and species name abbreviations shown are the same adopted in the genome annotation of each organism. Abbreviations of organism names are indicated in the legend of Figure 2. The numbers indicate the bootstrap values >50%.
Characterization of the imuAB/dnaE2 expression
The genomic organization of the CC3213 (imuA), CC3212 (imuB) and CC3211 (dnaE2) genes suggests that they may constitute an operon (Figure 1). In fact, in P.putida, Sinorhizobium meliloti and Agrobacterium tumefaciens, the orthologs of these genes are co-expressed in a single RNA (17). In order to evaluate the co-expression of these genes in C.crescentus, we designed RT–PCR experiments with the oligonucleotides represented in Figure 4A. The results confirmed that the three genes are expressed as a single operon, and the polycistronic RNA can be detected both in UV-irradiated and non-irradiated exponentially growing cells.
Figure 4 Expression of the imuAB dnaE2 operon. (A) Schematic diagram for this operon indicating the position of the oligonucleotides (small arrows) used in the RT–PCR strategy to detect the co-expression of the three genes. Template RNA was extracted from exponentially growing cells, both from non-irradiated cultures and from cultures irradiated with 60 J/m2 of UV and recovered after 45 min. The RT–PCR products were visualized after electrophoresis in 1% agarose gel stained with ethidium bromide. (B) Expression of uvrA and imuA promoters in transcriptional fusions with the lacZ gene. ?-Galactosidase activity was measured 90 min after irradiation with UV light, both in the wild-type and recA strains.
In -proteobacteria, the LexA-binding site has been characterized for many species. The typical sequence GTTCN7GTTC is apparently well conserved in all species examined to date (34,35). The operon containing imuA, imuB and dnaE2 presents this typical LexA-binding sequence 66 bp upstream from the imuA start codon, indicating that this operon is SOS regulated. In order to examine whether the expression of this operon is regulated in response to DNA damage, we cloned the promoter of imuA upstream the reporter gene lacZ in the pRKlacZ290 vector. For comparison, the promoter of the uvrA gene, which also bears a putative -proteobacteria type SOS box 78 bp upstream the start codon, was cloned in the same vector. Determination of ?-galactosidase activity was performed both for the wild-type NA1000 strain, and for a derivative recA deletion mutant, constructed in this work. Both promoters were strongly induced 90 min after UV irradiation in a dose-responsive fashion in the wild-type strain (Figure 4B), although the expression of lacZ under control of the imuAB/dnaE2 promoter is much higher than the one observed for the uvrA promoter.
In our conditions, optimal promoter expression was achieved 90–120 min after irradiation (data not shown), which is in good agreement with the previous reports of Weigle reactivation time course experiments in this organism (19). On the other hand, damage-induced activity of both promoters is completely abolished in a recA-deleted strain, indicating that both are regulated in a RecA-dependent manner. These results confirm the prediction that both uvrA and the imuAB/dnaE2 operon are controlled by the SOS regulon in C.crescentus.
The increased activity of the imuA promoter resulting in the induction of all three genes in the operon was further investigated by quantitative RT–PCR experiments. In addition, the expression of the dinB gene (encoded by the CC2466 ORF), which does not present an SOS box in its putative promoter region, was also tested. The results shown in Figure 5 demonstrate that imuA, imuB and dnaE2 expression levels are increased similarly in response to UV damage. Furthermore, the damage induction of the three genes is abolished in the recA strain, indicating that they are coordinately upregulated by the SOS response. The SOS box-containing promoter upstream of imuA presumably mediates this coordinated expression. In contrast, the expression of dinB remains unaffected by DNA damage, suggesting that this gene is not involved in damage tolerance in C.crescentus.
Figure 5 Quantitative RT–PCR analysis of UV-induced gene expression. RNAs were extracted from exponentially growing recA and wild-type cells before and 45 min after irradiation with 60 J/m2 UVC. Relative levels of expression shown represent the average of two independent experiments done in triplicate.
Genetic characterization of damage-inducible mutagenesis in C.crescentus
Several strains disrupted in the imuA, imuB and dnaE2 genes, as well as in the dinB gene, were constructed (Table 1). These gene disruptions had no effect upon cell growth (data not shown). Compared with the wild type, all the single mutant strains deficient in the genes of the operon were slightly sensitive to UV (Figure 6A). In contrast, the dinB strain showed wild-type levels of UV resistance, indicating that DinB is not involved in UV damage tolerance. The double mutant imuB dnaE2 showed no increased sensitivity to UV, indicating that the products of these genes act in the same tolerance pathway. A dinB imuB double mutant was also constructed, to evaluate if DinB could be acting in DNA damage tolerance in an imuAB-deficient background. The dinB mutation had no effect on the resistance to UV, further indicating that this gene is not involved in UV damage resistance.
Figure 6 Effects of imuAB, dnaE2 and dinB mutations on UV resistance and mutagenesis in C.crescentus. (A) Survival curves. (B) Frequency of Rifr mutations induced by UV in the rpoB gene. The results shown are the mean of at least five independent experiments done in triplicate. Error bars indicate the SD.
The mutagenesis of these strains was determined by the rifampicin resistance assay. This assay detects point mutations in the rpoB gene, which encodes the ?-subunit of bacterial RNA polymerase, and is very useful due to the high conservation of the target gene, allowing its utilization in a wide variety of bacterial species (36,37). Figure 6B shows the mutagenesis induced by UV light in all the strains. It is clear that dinB depletion did not affect UV-induced mutagenesis, in contrast to the disruption of imuA, imuB and dnaE2. The UV-induced mutations are almost equally diminished in these three strains, and reduced to approximately the same extent in the imuB dnaE2 double mutant. Altogether, the results shown in Figure 6 demonstrate that the imuAB dnaE2 operon is involved in mutagenic repair of UV lesions, and suggest that the proteins encoded by this operon cooperate in the same pathway, as evidenced by the similar phenotypes of the single mutants, and by the epistatic effect of imuB and dnaE2 mutations. The dinB imuB strain presents the same levels of UV-induced mutagenesis as the imuB single mutant, confirming that DinB is not implicated in error-prone repair in the conditions tested.
Similar experiments were performed to investigate the effects of the UV mimetic drug mit C in C.crescentus. The effects of mit C on the survival of the different strains are much more pronounced than those of UV light (Figure 7A). The single mutants imuA, imuB and dnaE2 mutants are extremely sensitive to this condition, revealing that this operon plays a major role in the tolerance to the mit C-induced DNA lesions. The imuB dnaE2 double mutant does not exhibit an exacerbated phenotype, indicating again that these genes cooperate in the same pathway of damage tolerance. On the other hand, the dinB mutant is as resistant as the wild type, and the dinB imuB double mutant shows no further sensitivity enhancement, demonstrating that DinB is not involved in the tolerance to mit C-induced DNA lesions. Both the UV and the mit C sensitivity phenotypes of imuA could be complemented by vector pMR3213, containing the wild-type imuA allele. Similarly, the imuB strain could be complemented by cosmid 3A4, which contains integral imuA and imuB genes, but lacking dnaE2 (data not shown). These results confirm that the phenotypes of imuA and imuB strains are not due to negative effects on the expression of downstream genes in the operon.
Figure 7 Effects of imuAB, dnaE2 and dinB mutations on mit C resistance and mutagenesis in C.crescentus. (A) Survival curves. (B) Frequency of Rifr mutations induced by mit C in the rpoB gene. The results shown are the mean of at least four independent experiments done in triplicate. Error bars indicate the SD.
The mutagenesis induced by mit C was also investigated using the rifampicin assay (Figure 7B). The results show a dramatic decrease in mit C mutagenesis in all strains devoid of genes belonging to the damage-inducible operon of C.crescentus. On the other hand, dinB is clearly not required at all for mit C mutagenesis. These data confirm that the imuAB/dnaE2 operon accounts for most of the mutagenic DNA repair in C.crescentus cells exposed to UV light and mit C.
Mutational specificity of UV light in C.crescentus
The nature of imuAB/dnaE2-dependent and independent mutations caused by UV light was analyzed in the rpoB gene. Rifampicin-resistant mutants obtained after UV radiation of the wild-type, dnaE2 and imuB strains were randomly selected for sequencing of the rpoB cluster II of rifampicin resistance (37), which contained the Rifr mutations in all mutants analyzed. The mutation spectra observed are shown in Figure 8. The mutation signature of UV light is remarkably similar in dnaE2 and imuB strains. G:C A:T transitions are by far the most common type of substitution, constituting 90% of the base changes in both strains (Table 2). On the other hand, UV mutagenesis in the parental strain is significantly modified to a high proportion of G:C C:G transversions and tandem substitutions concentrated in a hotspot consisting of a run of three Cs, which gives rise to only G:C A:T mutations in dnaE2 and imuB strains (Figure 8). This increase in transversions and tandem substitutions in the parental strain is accompanied by concomitant reduction in the fraction of G:C A:T transitions, as shown in Table 2. These other types of mutations are probably an outcome of the mutagenic activity of the imuAB/dnaE2 operon of C.crescentus, since they are not observed in the dnaE2 and imuB mutants. G:C C:G transversions are not a significant part of the mutagenic spectrum of UV light in E.coli (37,38). These results indicate that imuB and dnaE2 cooperate in a lesion bypass pathway, which has a different specificity than that involving umuDC in E.coli.
Figure 8 UV mutation spectra in the rpoB gene. Nucleotides 1557–1625, comprising the cluster II of rifampicin resistance, are shown in this figure. All mutants analyzed had mutations in this region of the rpoB gene. Mutations in shaded background correspond to repetitions of the same base substitution, and tandem mutations are underlined. The total number of independent mutations sequenced is indicated for each strain.
Table 2 Frequencies of base substitutions in the rpoB gene induced by UV light
DISCUSSION
In the present work, we demonstrate that a highly conserved gene expression unit consisting of the genes imuAB and dnaE2, controlled by the SOS regulon, is responsible for most of the damage-inducible mutagenesis in C.crescentus. The genes imuA and imuB encode previously uncharacterized proteins, which are shown here for the first time to be involved in DNA damage tolerance. These genes are present in several genomes where umuDC orthologs cannot be found, such as the -proteobacteria branch, and it is intuitive to argue that they may functionally replace these proteins in damage-inducible mutagenesis in a wide variety of bacterial species. It is very interesting to note that, independently of being organized as single transcriptional units or as independent genes, imuB and dnaE2 always co-exist in bacterial genomes, strongly suggesting that their activities are interconnected.
The data obtained here clearly indicate that this operon plays a major role in damage-induced mutagenesis in C.crescentus, especially after mit C exposure. The lack of mit C mutagenesis in imuAB/dnaE2-deficient strains correlates well with the extreme sensitivity they show to this agent. Although mit C mutagenesis is mostly dependent on these genes, at least one-third of the UV-induced mutations are still generated in the imu- or dnaE2-deficient genetic backgrounds. These mutations are not caused by the action of the dinB gene, as evidenced by the phenotypes of dinB and imuB dinB mutant strains. The nature of this mutagenic process remains to be elucidated. In E.coli, it has been shown that umu-independent UV mutagenesis may occur, depending on the experimental system used (39). Also, in B.subtilis, residual UV mutagenesis can be detected in varying levels, even in cells devoid of both Y-family DNA polymerases (40,41). It is interesting to note that no other protein belonging to the UmuC superfamily can be found in the C.crescentus genome, and neither there are polB orthologs. Thus, imuAB/dnaE2-independent UV mutagenesis might occur via a still uncharacterized pathway. Recently, it has been suggested that the dnaE gene of C.crescentus may also be a part of the SOS regulon (42). Overexpression of the replicative polymerase encoded by dnaE may account for the imuAB/dnaE2-independent UV mutagenesis. On the other hand, mit C mutagenesis shows an absolute dependence on the activity of the imuAB dnaE2 operon, indicating that DnaE and the ImuAB DnaE2 machinery may have different lesion bypass capacities.
The mutation spectra of UV-induced lesions in C.crescentus show a dramatic qualitative difference between wild-type and imuB or dnaE2 strains. G:C A:T transitions, a hallmark of UV mutagenesis in E.coli, are observed in the mutant strains, while wild-type cells also have a high proportion of tandem substitutions and G:C C:G transversions. This novel mutational signature, dependent on the activity of the imuAB/dnaE2 operon, probably reflects a difference in polymerase preference for nucleotide incorporation during TLS. On the other hand, the mutations induced by UV light in M.tuberculosis in a dnaE2-dependent manner are different from the observed in C.crescentus, resembling the UV mutagenic spectrum of E.coli (14). More detailed studies of both systems are still needed to clarify this discrepancy, although it is clear that there are fundamental differences between these organisms, since imuA orthologs are not present in M.tuberculosis and related bacteria. In addition, the use of the rpoB gene as target for mutational analysis has intrinsic limitations, such as the underestimation of frameshifts and the limited number of mutations detectable as gain of function alleles for Rifr. The development of more sophisticated genetic tools for the study of mutagenesis is mandatory for a deeper understanding of the basic mechanisms of this aspect of DNA metabolism in C.crescentus.
The DNA polymerase III holoenzyme is responsible for chromosomal replication. The -subunit of DNA polymerase III, encoded by the dnaE gene, is essential in E.coli (43). In C.crescentus, the involvement of the dnaE gene (ORF CC1926) in DNA replication has also been demonstrated previously (31). However, in M.tuberculosis, the deletion of dnaE2 causes no growth defect (14), similar to what was observed for C.crescentus. It was demonstrated years ago that dnaE is required for UV mutagenesis in E.coli (44,45). Currently, it is thought that successful TLS takes place after a blockage of Pol III and formation of a RecA filament on single-stranded DNA. Then, Y-family polymerases displace the Pol III catalytic subunit and mediate lesion bypass, synthesizing patches of DNA sufficiently sized to allow the hand-off of replication back to Pol III. These polymerase switches are mediated by interactions with the ?-clamp, the processivity factor of Pol III (46–48). The observation that several bacterial species bear a second copy of the dnaE gene, which seems to be solely devoted to DNA damage response, is very intriguing. The presence of this second, damage inducible, Pol III catalytic subunit may be an alternative manner some bacteria have found to control polymerase switch during lesion bypass. Additionally, DnaE2 may itself act as a TLS polymerase, as shown in B.subtilis and S.pyogenes (12,13).
In spite of the functional and phylogenetic characterization provided in this work, the actual biochemical functions of imuA and imuB genes remain to be elucidated. However, the genetic data shown here indicate that these genes cooperate with dnaE2 in an error-prone repair pathway. ImuB is a member of the UmuC superfamily, and it is tempting to speculate about a model in which ImuB and DnaE2 would act in concert, as shown for Y-family polymerases and Pol III in TLS in E.coli. This hypothesis is consistent with the genetic requirement for both imuB and dnaE2 in damage-inducible mutagenesis shown in this work, and is further supported by the epistatic effect of these gene disruptions. This model would present a functional conservation of the overall TLS scenario in bacteria, although natural selection has chosen different role-players within different domains of the bacterial kingdom. Shedding light on the reasons why different bacteria should have distinct mutagenic repair machinery is a challenging question, which deserves further examination.
SUPPLEMENTARY MATERIAL
Supplementary Material is available at NAR Online.
ACKNOWLEDGEMENTS
The authors would like to thank Dr Craig Stephens (Santa Clara University) for the gift of the 3A4 cosmid. Financial support was obtained from FAPESP (S?o Paulo, Brazil) and CNPq (Brasília, Brazil). R.S.G. is a Postdoctoral fellow from FAPESP. Funding to pay the Open Access publication charges for this article was provided by FAPESP.
REFERENCES
Rangarajan, S., Woodgate, R., Goodman, M.F. (1999) A phenotype for enigmatic DNA polymerase II: a pivotal role for pol II in replication restart in UV-irradiated Escherichia coli Proc. Natl Acad. Sci. USA, 96, 9224–9229 .
Napolitano, R., Janel-Bintz, R., Wagner, J., Fuchs, R.P. (2000) All three SOS-inducible DNA polymerases (Pol II, Pol IV and Pol V) are involved in induced mutagenesis EMBO J., 19, 6259–6265 .
Wagner, J., Gruz, P., Kim, S.R., Yamada, M., Matsui, K., Fuchs, R.P., Nohmi, T. (1999) The dinB gene encodes a novel E.coli DNA polymerase, DNA pol IV, involved in mutagenesis Mol. Cell, 4, 281–286 .
Ohmori, H., Friedberg, E., Fuchs, R.P., Goodman, M.F., Hanaoka, F., Hinkle, D., Kunkel, T.A., Lawrence, C.W., Livneh, Z., Nohmi, T. (2001) The Y-family of DNA polymerases Mol. Cell, 8, 7–8 .
Kim, S.R., Maenhaut-Michel, G., Yamada, M., Yamamoto, Y., Matsui, K., Sofuni, T., Nohmi, T., Ohmori, H. (1997) Multiple pathways for SOS-induced mutagenesis in Escherichia coli: an overexpression of dinB/dinP results in strongly enhancing mutagenesis in the absence of any exogenous treatment to damage DNA Proc. Natl Acad. Sci. USA, 94, 13792–13797 .
Hersh, M.N., Ponder, R.G., Hastings, P.J., Rosenberg, S.M. (2004) Adaptive mutation and amplification in Escherichia coli: two pathways of genome adaptation under stress Res. Microbiol., 155, 352–359 .
Elledge, S.J. and Walker, G.C. (1983) Proteins required for ultraviolet light and chemical mutagenesis: identification of the products of the umuC locus of Escherichia coli J. Mol. Biol., 164, 175–192 .
Shinagawa, H., Kato, T., Ise, T., Makino, K., Nakata, A. (1983) Cloning and characterization of the umu operon responsible for inducible mutagenesis in Escherichia coli Gene, 23, 167–174 .
Reuven, N.B., Arad, G., Maor-Shoshani, A., Livneh, Z. (1999) The mutagenesis protein UmuC is a DNA polymerase activated by UmuD', RecA, and SSB and is specialized for translesion replication J. Biol. Chem., 274, 31763–31766 .
Tang, M., Shen, X., Frank, E.G., O'Donnell, M., Woodgate, R., Goodman, M.F. (1999) UmuD'(2)C is an error-prone DNA polymerase, Escherichia coli pol V Proc. Natl Acad. Sci. USA, 96, 8919–8924 .
Tippin, B., Pham, P., Goodman, M.F. (2004) Error-prone replication for better or worse Trends Microbiol., 12, 288–295 .
Bruck, I., Goodman, M.F., O'Donnell, M. (2003) The essential C family DnaE polymerase is error-prone and efficient at lesion bypass J. Biol. Chem., 278, 44361–44368 .
Le Chatelier, E., Bécherel, O.J., d'Alen?on, E., Canceill, D., Ehrlich, S.D., Fuchs, R.P., Jannière, L. (2004) Involvement of DnaE, the second replicative DNA polymerase from Bacillus subtilis, in DNA mutagenesis J. Biol. Chem., 279, 1757–1767 .
Boshoff, H.I., Reed, M.B., Barry, C.E., III, Mizrahi, V. (2003) DnaE2 polymerase contributes to in vivo survival and the emergence of drug resistance in Mycobacterium tuberculosis Cell, 113, 183–193 .
Davis, E.O., Dullaghan, E.M., Rand, L. (2002) Definition of the mycobacterial SOS box and use to identify LexA-regulated genes in Mycobacterium tuberculosis J. Bacteriol., 184, 3287–3295 .
Rand, L., Hinds, J., Springer, B., Sander, P., Buxton, R.S., Davis, E.O. (2003) The majority of inducible DNA repair genes in Mycobacterium tuberculosis are induced independently of RecA Mol. Microbiol., 50, 1031–1042 .
Abella, M., Erill, I., Jara, M., Mazon, G., Campoy, S., Barbe, J. (2004) Widespread distribution of a lexA-regulated DNA damage-inducible multiple gene cassette in the Proteobacteria phylum Mol. Microbiol., 54, 212–222 .
Nierman, W.C., Feldblyum, T.V., Laub, M.T., Paulsen, I.T., Nelson, K.E., Eisen, J.A., Heidelberg, J.F., Alley, M.R., Ohta, N., Maddock, J.R., et al. (2001) Complete genome sequence of Caulobacter crescentus Proc. Natl Acad. Sci. USA, 98, 4136–4141 .
Bender, R.A. (1984) Ultraviolet mutagenesis and inducible DNA repair in Caulobacter crescentus Mol. Gen. Genet., 197, 399–402 .
Ely, B. (1991) Genetics of Caulobacter crescentus Methods Enzymol., 204, 372–384 .
Prentki, P. and Krisch, H.M. (1984) In vitro insertional mutagenesis with a selectable DNA fragment Gene, 29, 303–313 .
Miller, J.H. A Short Course in Bacterial Genetics—A Laboratory Manual and Handbook for Escherichia Coli and Related Bacteria, (1992) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press pp. 72–74 .
Pfaffl, M.W. (2001) A new mathematical model for relative quantification in real-time RT–PCR Nucleic Acids Res., 29, e45 .
Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res., 25, 3389–3402 .
Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G. (1997) The ClustalX Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools Nucleic Acids Res., 25, 4876–4882 .
Nicholas, K.B. and Nicholas, H.B., Jr. (1997) GeneDoc: a tool for editing and annotating multiple sequence alignments .
Felsenstein, J. (1989) PHYLIP Phylogeny Inference Package (Version 3.2) Cladistics, 5, 164–166 .
Saitou, N. and Nei, M. (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees Mol. Biol. Evol., 4, 406–425 .
Page, R.D.M. (1996) TREEVIEW: an application to display phylogenetic trees on personal computers Comput. Appl. Biosci., 12, 357–359 .
Martins-Pinheiro, M., Galhardo, R.S., Lage, C., Lima-Bessa, K.M., Aires, K.A., Menck, C.F. (2004) Different patterns of evolution for duplicated DNA repair genes in bacteria of the Xanthomonadales group BMC Evol. Biol., 4, 29 .
Lo, T., van Der Schalie, E., Werner, T., Brun, Y.V., Din, N. (2004) A temperature-sensitive mutation in the dnaE gene of Caulobacter crescentus that prevents initiation of DNA replication but not ongoing elongation of DNA J. Bacteriol., 186, 1205–1212 .
Weinel, C., Nelson, K.E., Tümmier, B. (2002) Global features of the Pseudomonas putida KT2440 genome sequence Environ. Microbiol., 4, 809–818 .
Bruggemann, H., Henne, A., Hoster, F., Liesegang, H., Wiezer, A., Strittmatter, A., Hujer, S., Durre, P., Gottschalk, G. (2004) The complete genome sequence of Propionibacterium acnes, a commensal of human skin Science, 305, 671–673 .
Fernandez de Henestrosa, A.R., Rivera, E., Tapias, A., Barbe, J. (1998) Identification of the Rhodobacter sphaeroides SOS box Mol. Microbiol., 28, 991–1003 .
Tapias, A. and Barbe, J. (1999) Regulation of divergent transcription from the uvrA-ssb promoters in Sinorhizobium meliloti Mol. Gen. Genet., 262, 121–130 .
Kim, M., Wolff, E., Huang, T., Garibyan, L., Earl, A.M., Battista, J.R., Miller, J.H. (2004) Developing a genetic system in Deinococcus radiodurans for analyzing mutations Genetics, 166, 661–668 .
Garibyan, L., Huang, T., Kim, M., Wolff, E., Nguyen, A., Nguyen, T., Diep, A., Hu, K., Iverson, A., Yang, H., et al. (2003) Use of the rpoB gene to determine the specificity of base substitution mutations on the Escherichia coli chromosome DNA Repair (Amst.), 2, 593–608 .
Livneh, Z., Cohen-Fix, O., Skaliter, R., Elizur, T. (1993) Replication of damaged DNA and the molecular mechanism of ultraviolet light mutagenesis Crit. Rev. Biochem. Mol. Biol., 28, 465–513 .
Christensen, J.R., LeClerc, J.E., Tata, P.V., Christensen, R.B., Lawrence, C.W. (1988) UmuC function is not essential for the production of all targeted lacI mutations induced by ultraviolet light J. Mol. Biol., 203, 635–641 .
Sung, H.M., Yeamans, G., Ross, C.A., Yasbin, R.E. (2003) Roles of YqjH and YqjW, homologs of the Escherichia coli UmuC/DinB or Y superfamily of DNA polymerases, in stationary-phase mutagenesis and UV-induced mutagenesis of Bacillus subtilis J. Bacteriol., 185, 2153–2160 .
Duigou, S., Ehrlich, S.D., Noirot, P., Noirot-Gros, M.F. (2004) Distinctive genetic features exhibited by the Y-family DNA polymerases in Bacillus subtilis Mol. Microbiol., 54, 439–451 .
Erill, I., Jará, M., Salvador, N., Escribano, M., Campoy, S., Barbe, J. (2004) Differences in LexA regulon structure among Proteobacteria through in vivo assisted comparative genomics Nucleic Acids Res., 32, 6617–6626 .
McHenry, C.S. (1998) DNA polymerase III holoenzyme of Escherichia coli Annu. Rev. Biochem, 57, 519–550 .
Bridges, B.A. and Bates, H. (1990) Mutagenic DNA repair in Escherichia coli. XVIII. Involvement of DNA polymerase III alpha-subunit (DnaE protein) in mutagenesis after exposure to UV light Mutagenesis, 5, 35–38 .
Bridges, B.A. and Mottershead, R.P. (1976) Mutagenic DNA repair in Escherichia coli. III. Requirement for a function of DNA polymerase III in ultraviolet-light mutagenesis Mol. Gen. Genet., 144, 53–58 .
Sutton, M.D., Opperman, T., Walker, G.C. (1999) The Escherichia coli SOS mutagenesis proteins UmuD and UmuD' interact physically with the replicative DNA polymerase Proc. Natl Acad. Sci. USA, 96, 12373–12378 .
Bunting, K.A., Roe, S.M., Pearl, L.H. (2003) Structural basis for recruitment of translesion DNA polymerase PolIV/DinB to the beta-clamp EMBO J., 22, 5883–5892 .
Fujii, S. and Fuchs, R.P. (2004) Defining the position of switches between replicative and bypass DNA polymerases EMBO J., 23, 4342–4352 .
Evinger, M. and Agabian, N. (1977) Envelope-associated nucleoid from Caulobacter crescentus stalked and swarmer cells J. Bacteriol., 132, 294–301 .
Simon, R., Prieffer, U., Puhler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria Biotechnology, 1, 784–790 .
Tsai, J.W. and Alley, M.R. (2000) Proteolysis of the McpA chemoreceptor does not require the Caulobacter major chemotaxis operon J. Bacteriol., 182, 504–507 .
Roberts, R.C., Toochinda, C., Avedissian, M., Baldini, R.L., Gomes, S.L., Shapiro, L. (1996) Identification of a Caulobacter crescentus operon encoding hrcA, involved in negatively regulating heat-inducible transcription and the chaperone gene grpE J. Bacteriol., 178, 1829–1841 .
Gober, J.W. and Shapiro, L. (1992) A developmentally regulated Caulobacter flagellar promoter is activated by 3' enhancer and IHF binding elements Mol. Biol. Cell, 3, 913–916 .
Alley, M.R., Gomes, S.L., Alexander, W., Shapiro, L. (1991) Genetic analysis of a temporally transcribed chemotaxis gene cluster in Caulobacter crescentus Genetics, 129, 333–341 .(Rodrigo S. Galhardo, Raquel P. Rocha, Ma)